From: Jeff Moyer Subject: [PATCH 2/6] jbd: yield the device queue when waiting for commits Date: Fri, 2 Jul 2010 15:58:15 -0400 Message-ID: <1278100699-24132-3-git-send-email-jmoyer@redhat.com> References: <1278100699-24132-1-git-send-email-jmoyer@redhat.com> Cc: axboe@kernel.dk, linux-kernel@vger.kernel.org, vgoyal@redhat.com, tao.ma@oracle.com, Jeff Moyer To: linux-ext4@vger.kernel.org Return-path: Received: from mx1.redhat.com ([209.132.183.28]:2840 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756777Ab0GBT6c (ORCPT ); Fri, 2 Jul 2010 15:58:32 -0400 In-Reply-To: <1278100699-24132-1-git-send-email-jmoyer@redhat.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: This patch gets CFQ back in line with deadline for iozone runs, especially those testing small files + fsync timings. Signed-off-by: Jeff Moyer --- fs/jbd/journal.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-) diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index 93d1e47..7bf7528 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include @@ -183,6 +184,7 @@ loop: should_sleep = 0; if (should_sleep) { spin_unlock(&journal->j_state_lock); + blk_yield(journal->j_dev->bd_disk->queue, NULL); schedule(); spin_lock(&journal->j_state_lock); } @@ -549,6 +551,11 @@ int log_wait_commit(journal_t *journal, tid_t tid) while (tid_gt(tid, journal->j_commit_sequence)) { jbd_debug(1, "JBD: want %d, j_commit_sequence=%d\n", tid, journal->j_commit_sequence); + /* + * Give up our I/O scheduler time slice to allow the journal + * thread to issue I/O. + */ + blk_yield(journal->j_dev->bd_disk->queue, journal->j_task); wake_up(&journal->j_wait_commit); spin_unlock(&journal->j_state_lock); wait_event(journal->j_wait_done_commit, -- 1.6.5.2